Autism-like features in individuals at clinical high risk for psychosis: a longitudinal research using the PANSS Autism Severity Score (PAUSS).

Purpose: The PANSS Autism Severity Score (PAUSS) is a widely used measure of autism-like features in early psychosis, including mental states at “Clinical High Risk for Psychosis” (CHR-P). However, evidence regarding its relevance to long-term outcomes and treatment response is very scarce. Thus, the main aim of this investigation was to compare baseline clinical characteristics and longitudinal outcomes/treatment response between CHR-P subjects with or without autistic features recruited into a specialist early intervention service during a 2-year follow-up period. Methods: CHR-P participants completed the Social and Occupational Functioning Assessment Scale (SOFAS) and the Positive And Negative Syndrome Scale (PANSS) at baseline and over the follow-up. Kaplan-Meyer survival analysis, mixed-design ANOVA, and binary logistic regression analysis were performed. Results: Thirty-six (16.8%) of the 214 enrolled CHR-P individuals scored above the PAUSS cut-off of 30 (PAUSS + subgroup). At baseline, they showed higher PANSS and SOFAS scores, as well as higher prevalence rates of substance abuse and individuals not engaged in education, employment, or training. Over the follow-up period, they showed worse clinical and functional outcomes, including higher incidence rates of hospital re-admission (28.7% vs. 11.6%; Hazard Ratio [HR] = 3.38) and conversion to psychosis (26.0% vs. 12.0%; HR = 2.71), as well as lower likelihood of both functional (Hazard Risk [HR] = 4.23) and symptomatic (HR = 6.50) remission. However, our specialist intervention has proven effective in improving psychopathological and functional parameters over time even in the PAUSS + subgroup. Conclusions: Our findings suggest that the PAUSS specifically identifies a CHR-P subgroup with greater clinical and functional severity at presentation and worse longitudinal outcomes.